Replace stack/mask/reduce by indexing in _hsv_to_rgb

Fixes #7753

torchvision/transforms/v2/functional/_color.py

@@ -325,14 +325,22 @@ def _hsv_to_rgb(img: torch.Tensor) -> torch.Tensor:
     p = one_minus_s.mul_(v).clamp_(0.0, 1.0)
     i.remainder_(6)
 
-    mask = i.unsqueeze(dim=-3) == torch.arange(6, device=i.device).view(-1, 1, 1)
-
-    a1 = torch.stack((v, q, p, p, t, v), dim=-3)
-    a2 = torch.stack((t, v, v, q, p, p), dim=-3)
-    a3 = torch.stack((p, p, t, v, v, q), dim=-3)
-    a4 = torch.stack((a1, a2, a3), dim=-4)
-
-    return (a4.mul_(mask.unsqueeze(dim=-4))).sum(dim=-3)
+    vpqt = torch.stack((v, p, q, t), dim=-3)
+
+    # vpqt -> rgb mapping based on i
+    select = torch.tensor([[0, 2, 1, 1, 3, 0], [3, 0, 0, 2, 1, 1], [1, 1, 3, 0, 0, 2]], dtype=torch.long)
+    if img.device.type == "cuda":
+        select = select.pin_memory(img.device)
+    select = select.to(device=img.device, non_blocking=True)
+
+    select = select[:, i]
+    if select.ndim > 3:
+        # if input.shape is (B, ..., C, H, W) then
+        # select.shape is (C, B, ...,  H, W)
+        # thus we move C axis to get (B, ..., C, H, W)
+        select = select.moveaxis(0, -3)
+
+    return vpqt.gather(-3, select)
 
 
 def adjust_hue_image_tensor(image: torch.Tensor, hue_factor: float) -> torch.Tensor: